Prev: Can anyone recommend a JavaScript Tree with drag and drop functionality?
Next: "heredoc" in javascript
From: Gregor Kofler on 11 Feb 2010 16:16
Jorge meinte:

> I fully understand the way my toaster works, my freezer, the washing
> machine, a Scotch Brite��, a bicycle, even my car, a C compiler, a
> truetype font, a JS minifier, a JSON parser, a code128 barcode, the
> gzip algorithm, base64 encoding, an http server, a serial port, a
> digital clock, a keyboard, a microprocessor, a switching-mode DC power
> supply... etc.(*)

I know how nuclear fission or fusion works - yet don't have my own small
reactor at home, since I lack some tools needed.

I do have a text editor and browser, though.

> So what ???

....are you trying to "prove"?

Gregor


--
http://www.gregorkofler.com
From: Thomas 'PointedEars' Lahn on 12 Feb 2010 20:29
mscir wrote:

> David Mark wrote:
>> mscir wrote:
>>> The second hit included a perfect working example that is free.
>>
>> http://www.dhtmlgoodies.com/index.html?whichScript=drag-drop-folder-tree
>>
>> It's free, but hasn't been updated since 2006. That could mean one of
>> two things. It's such a perfect cross-browser rendition that it needs
>> no updates (highly unlikely) or it is abandoned. I didn't look into it,
>> but it is always safest to assume that these things are bunkware. ;)
>
> http://www.dhtmlgoodies.com/scripts/drag-drop-folder-tree/drag-drop
> folder-tree.html
>
> This one works in my XP browsers:
> IE 8
> Firefox 3.6
> Opera 10
> Safari 4.0.4

That does not mean anything, particularly as you have only tested with one
operating system on one platform.

> Why I thought it was good enough to post here was: it's free,

True, it is free software (GPL v2.1+ licensed), and it can be downloaded
for free. However, free advice is seldom cheap. That goes especially for
scripts downloadable from self-proclaimed portal sites like
dhtmlgoodies.com.

> it works at least some of the time -

If that is a sufficient criterion for you for good software quality, you
should better stop developing (and recommending) software.

> even in late version browsers, and the OP can start learning about how
> it's done by downloading the code

No, they won't.

> and seeing which browsers break it. If he run into problems he can post
> about the problems here and learn more javascript, or learn enough to
> recognize better written examples of drag and drop when he does further
> research.

Having to unlearn a load of utter nonsense beforehand.

> So I thought this was a good recommendation, since he didn't specify
> stringent requirements.

But it was not. In fact, given the lack of specific requirements, the
requirements are even greater for a distributed application like this,
because it must not break even when exposed to yet unknown environments.

The problem with scripts poorly written such as this is that the OP is not
going to find any problems with it (like you) until it is too late, but
their users most certainly will. Starting with, ISTM, the users that do
not have sufficient client-side script support available as the markup it
is based on goes like this:

<ul>
<li id="node2" noDelete="true"><a href="#">Norway</a>
<ul>
<li id="node3" noRename="true"><a href="#">Stavanger</a></li>
<li id="node6"><a href="#">Bergen</a></li>
<li id="node7"><a href="#">Oslo</a></li>
</ul>
</li>
...

Search engines will find something to index, but nothing to follow here,
too. Great idea. NOT.

And they even dare declare this junk HTML 4.01 (Strict). This script is --
literally -- not worth one bit of the about 44 KiB it requires to "work".
And that is only a rough estimate.


PointedEars
--
var bugRiddenCrashPronePieceOfJunk = (
navigator.userAgent.indexOf('MSIE 5') != -1
&& navigator.userAgent.indexOf('Mac') != -1
) // Plone, register_function.js:16
From: Thomas 'PointedEars' Lahn on 12 Feb 2010 21:11
Thomas 'PointedEars' Lahn wrote:

> [...]
> The problem with scripts poorly written such as this is that the OP is
> not going to find any problems with it (like you) until it is too late,
> but their users most certainly will. Starting with, ISTM, the users that
> do not have sufficient client-side script support available as the markup
> it is based on goes like this:
>
> <ul>
> <li id="node2" noDelete="true"><a href="#">Norway</a>
> <ul>
> <li id="node3" noRename="true"><a href="#">Stavanger</a></li>
> <li id="node6"><a href="#">Bergen</a></li>
> <li id="node7"><a href="#">Oslo</a></li>
> </ul>
> </li>
> ...
>
> Search engines will find something to index, but nothing to follow here,
> too. Great idea. NOT.

OK, after digging through most of the script code it does not appear as if
this collection of scripts would set `click' listeners dynamically, like so
many others, so I can accept this as (poor) example code. (But would the
OP have recognized this?)

However, the scripts are based on invalid markup, they augment host objects
and use setAttribute() to create invalid attributes on which values they
rely, for which they should be dumped already. And then there are bogus
feature tests like `if (document.all) e = event;', mentioning of classes in
a prototype-based language, and too many globals (some of them undeclared),
to name a few more mistakes.

> And they even dare declare this junk HTML 4.01 (Strict). This script is
> -- literally -- not worth one bit of the about 44 KiB it requires to
> "work". And that is only a rough estimate.

So I stand by my initial recommendation: Do not use these scripts.


PointedEars
--
realism: HTML 4.01 Strict
evangelism: XHTML 1.0 Strict
madness: XHTML 1.1 as application/xhtml+xml
-- Bjoern Hoehrmann
From: Evertjan. on 18 Feb 2010 14:20
Laser Lips wrote on 18 feb 2010 in comp.lang.javascript:

>> As long as my English is better than your Dutch,
>> you should not complain in this international NG, methinks.
>
> Sorry Evertjan, was just being Pedantic.

Wel taken. What is a NG without emotions?

> Well, I managed to create my own script which has drag and drop as
> well as rules you can set for nodes which can and can't be dragged on
> certain parents. Also has the ability to reorder leaf nodes.
>
> Yes, I did learn that to make a good tree script. It's logical to use
> nested unordered lists.

Hopefully you will be able to expand it in a year or two without having to
secandguess your present programming ideas. [My probelm at present]

In any case it is much easier than the hidden ideas of a third [second?]
party.

--
Evertjan.
The Netherlands.
(Please change the x'es to dots in my emailaddress)
First  |  Prev  | 
Pages: 1 2 3
Prev: Can anyone recommend a JavaScript Tree with drag and drop functionality?
Next: "heredoc" in javascript